.png){kind=icon}
返回
Distributed Bayesian Inference Over Sensor Networks
.png){kind=icon}
PRE
AI
DOI：10.1109/TCYB.2021.3106660.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
In this article, two novel distributed variational Bayesian (VB) algorithms for a general class of conjugate-exponential models are proposed over synchronous and asynchronous sensor networks. First, we design a penalty-based distributed VB (PB-DVB) algorithm for synchronous networks, where a penalty function based on the Kullback-Leibler (KL) divergence is introduced to penalize the difference of posterior distributions between nodes. Then, a token-passing-based distributed VB (TPB-DVB) algorithm is developed for asynchronous networks by borrowing the token-passing approach and the stochastic variational inference. Finally, applications of the proposed algorithm on the Gaussian mixture model (GMM) are exhibited. Simulation results show that the PB-DVB algorithm has good performance in the aspects of estimation/inference ability, robustness against initialization, and convergence speed, and the TPB-DVB algorithm is superior to existing token-passing-based distributed clustering algorithms.
Keyword:
Inference algorithms
Clustering algorithms
Convergence
Bayes methods
Hidden Markov models
Biological system modeling
Approximation algorithms
Conjugate-exponential model
distributed variational Bayesian (VB)
Kullback-Leibler (KL) divergence
sensor networks
token-passing method
期刊
IF:
10.5
论文数:
1.1W
被引数:
5.0W
机构
引用论文
A Fast Distributed Variational Bayesian Filtering for Multisensor LTV System With Non-Gaussian Noise
PREAI
分享收藏分享收藏Comparison of Superchilling and Supercooling on Extending the Fresh Quality of Beef Loin.png)
Foods
OAAI
FoodsIF0
OAAI分享收藏分享收藏
查看更多内容